Abstract: For the deficiency of current digital watermarking methods for English texts, the paper proposes a multiple text watermarking method which is based on the characters’ attribute and the words’ content. This method can make a huge contain of the watermarking, on the other hand, the watermarking can be hided effectively. Moreover, it has a strong ability of testing and correcting the watermarking after the demolishing made to the text content or the watermarking by attackers. Attacking experiments have shown that the watermarking has a good ability of robustness.

Key words: Text digital watermarking, Contain of watermarking, Integrality

摘要： 针对现有英文文本数字水印方法存在的不足，提出并实现了一种基于字符属性及单词内容的多重文本数字水印算法。该方法可以实现较大的水印容量且水印的隐蔽性好，同时针对攻击者对文本内容或水印的破坏，该算法具有较好的检测和纠错提取性能。攻击实验数据证明水印的鲁棒性较强。

关键词: 文本数字水印, 水印容量, 完整性